Adult Celiac Disease: What Is the Role of MRI?

Transcription

1 JOURNAL OF MAGNETIC RESONANCE IMAGING 24: (2006) Original Research Adult Celiac Disease: What Is the Role of MRI? Ernesto Tomei, MD, 1 Richard C. Semelka MD, 2 * Larissa Braga, MD, PhD, 2 Andrea Laghi, MD, 3 Pasquale Paolantonio, MD, 3 Mario Marini, MD, 1 Roberto Passariello, MD, 1 Marco Di Tola, MD, PhD, 4 Luigi Sabbatella, MD, PhD, 4 and Antonio Picarelli MD 4 Purpose: To evaluate the ability of MRI to identify intraand extraintestinal findings of celiac disease in an adult population. Materials and Methods: Forty-one subjects (18 men and 23 women; mean age 41.3 years; 31 with biopsy-proven celiac disease, and 10 healthy volunteers) underwent MRI of the small bowel. MR studies were performed on a 1.5-T magnet using T2-weighted half-fourier single-shot turbo spin-echo (HASTE) and true fast imaging in steady-state precession (True-FISP) sequences. The MR features and sensitivity, and the specificity and accuracy of some of these features are described. Results: In the 31 celiac patients, MRI showed bowel dilatation in 61.3% (N 19), increased number of ileal folds in 48.4% (N 15), reversed fold pattern abnormality in 38.7% (N 12), increased wall thickness in 16.1% (N 5), duodenal stenosis in 6.5% (N 2), intussusception in 12.9% (N 4), mesenteric lymphadenopathy in 41.9% (N 13), mesenteric vascular changes in 22.6% (N 7), ascites in 6.5% (N 2), and no abnormalities in 12.9% (N 4). The volunteers had unremarkable exams. The overall specificity and accuracy were 100%, and sensitivity was 79% and 75% for increased number of ileal folders and reversed fold pattern abnormality, respectively. Conclusion: MRI is able to demonstrate intra- and extraintestinal features that may lead to the diagnosis of celiac disease in adults. Key Words: MRI; celiac disease; small intestine; adults MRI; bowel disease J. Magn. Reson. Imaging 2006;24: Wiley-Liss, Inc. 1 Department of Radiological Sciences, University of Rome La Sapienza, Rome, Italy. 2 Department of Radiology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA. 3 Department of Radiological Sciences, University of Rome La Sapienza, Latina, Italy. 4 Department of Clinical Sciences, University of Rome La Sapienza, Rome, Italy. *Address reprint requests to: R.C.S., Department of Radiology, University of North Carolina at Chapel Hill, 101 Manning Drive, CB# Old Clinic Bldg., Chapel Hill, NC richsem@med.unc.edu Received July 4, 2005; Accepted May 11, DOI /jmri Published online 3 August 2006 in Wiley InterScience ( wiley.com). CELIAC DISEASE or gluten-sensitive enteropathy is a chronic disease that affects both children and adults. Some authors consider that celiac disease may be more prevalent than reported, and describe the reported cases as the tip of the celiac iceberg (1). In recent years the diagnosis of celiac disease has been established in greater frequency in the adult population, most likely reflecting the heightened awareness of clinicians and endoscopists (2 4). The diagnosis of adult celiac disease is challenging because of the wide range of clinical manifestations and their lack of specificity (5). The paucity of reports in the radiology literature suggests that there may be insufficient awareness of the potential role of abdominal imaging in establishing this diagnosis. The ability to recognize MR features in celiac patients is crucial since cross-sectional studies of the abdomen are frequently requested in cases of nonspecific abdominal symptoms. The role of enteroclysis in the diagnosis of celiac disease was discussed in previous studies (6,7), and a combination of ultrasound findings in adult celiac disease has been reported (8). However, the characteristic imaging findings of jejunoileal fold pattern reversal and jejunization of the ileum have not been revealed by ultrasound. Previous reports described the CT features of celiac disease, such as jejunoileal fold pattern reversal, small bowel intussusception, and benign mesenteric lymphadenopathy (9 13). Laghi et al (14) recently reported preliminary MR studies that described features of celiac disease in children and adults. The purpose of the current study was to describe intra- and extraintestinal MR findings in adult patients in the setting of celiac disease, and to determine whether normal subjects can be distinguished from patients with this disease. MATERIALS AND METHODS Patients Forty-one subjects (13 men and 18 women, mean age 41.3 years) were recruited for this study. Thirty-one of these subjects were patients who attended the celiac disease clinic and had a clinical history and histopathologically proven disease. The control group consisted of 10 healthy volunteers (five men and five women, mean age 38.7 years) with no history of gastrointestinal 2006 Wiley-Liss, Inc. 625

2 626 Tomei et al. disorders or abdominal surgery. Approval from the institutional review board and written consent from all of the subjects were obtained for this prospective analysis. MR studies were performed at the discretion of the clinicians responsible for patient management, with no involvement by the authors of this study. MRI Following an overnight fast, the patients ingested 1000 ml of contrast (polyethylene glycol (PEG) solution (SELG, Promefarm, Italy)) prior to the examination. Luminal distension was performed to avoid a collapsed bowel simulating a thickened bowel wall (pseudo wall thickening) (15). No other bowel preparation or antispasmodic was given prior to the study. MR images were acquired every five minutes for up to 30 minutes, or until the contrast reached the cecum. The average time for image analysis was 20 minutes. Studies were performed on a 1.5-T scanner (Magnetom Vision Plus; Siemens Medical Systems, Erlangen, Germany) using a phased-array body coil. The MR protocol included coronal and axial T2-weighted half-fourier single-shot turbo spin-echo (HASTE) sequences (TR infinity; TE 90 msec; echo train length 104; section thickness 6 7 mm; intersection gap 1.3 mm; FOV mm; matrix ), and coronal true fast imaging in steady-state precession (True-FISP) sequence (TR 4.8 msec; TE 2.3 msec; flip angle 50 ; section thickness 5 mm; no gap; FOV 380; matrix ). Imaging Analysis Images were reviewed by two board-certified radiologists who specialized in gastrointestinal imaging and were blinded to the patients clinical history and histopathological results. Agreement was reached by consensus. The MR features were described as follows: I) bowel dilatation, when at least three segments of bowel measured more than 3 cm in diameter, in the absence of distal stenosis (16); II) fold pattern abnormalities, when the number of jejunal folds was decreased (less than five folds per inch) and the number of ileal folds was increased, i.e., jejunization (more than five folds per inch) (7), reversal of the jejunoileal fold pattern was considered when both previous features were present (17); III) bowel wall thickening, when the wall measured more than 4 mm (18); IV) intestinal stenosis, when a narrowed intestinal segment was present, possibly in association with prestenotic dilatation with a transition zone (16); V) intestinal intussusception, when the appearance of bowel-within-bowel was visualized (19,20); V) enlarged mesenteric lymph nodes, when nodes measured more than 1 cm in diameter in the short axis (12); VI) mesenteric vascular changes, when dilated vessels were appreciated in the mesentery; and VII) ascites. Figure 1. Coronal T2-weighted HASTE image in a 30-year-old volunteer showing a normal fold pattern of jejunum (arrow) and normal bowel wall of an ileal loop (arrowhead); note also the normal small intestine caliber. Clinical, Laboratory, and Histopathological Analyses All of the patients with celiac disease had a clinical history of diarrhea and abdominal pain. Laboratory results showed anemia associated with positive antiendomysial antibody (EMA) measured by indirect immunofluorescence analysis (IFA). Biopsy was performed in all patients. Statistical Analysis We assessed the sensitivity, specificity, and accuracy of MRI for evaluating the presence of an increased number of ileal folds and a reversed fold pattern abnormality in the 41 subjects (31 diseased and 10 healthy). RESULTS All subjects in the control group were reported to have an unremarkable MR examination (Fig. 1). In the 31 patients with biopsy-proven celiac disease, MRI demonstrated bowel dilatation in 61.3% (19/31), increased number of ileal folds in 48.4% (15/31; Fig. 2), reversed fold pattern abnormality in 38.7% (12/31; Fig. 3), increased wall thickness in 16.1% (5/31; Fig. 4), duodenum stenosis in 6.5% (2/31; Fig. 5), intussusception in 12.9% (4/31; see Fig. 3), mesenteric lymphadenopathy in 41.9% (13/31; Fig. 6), mesenteric vascular changes in 22.6% (7/31; see Fig. 6), small volume ascites in 6.5% (2/31), and no abnormalities in 12.9% (4/31). None of the patients exhibited complete loss of intestinal folds.

3 MRI in Adult Celiac Disease 627 Figure 2. Coronal True-FISP in a 42-year-old patient with celiac disease that demonstrates an increased number of ileal folds (jejunization) and a normal fold pattern in the jejunum. Biopsy was consistent with mucosal partial atrophy in 54.8% (17/31) of the patients, and flat mucosa, i.e., complete atrophy, in 45.2% (14/31) of the patients. Biopsy ruled out malignancy in both patients with duodenal stenosis. Regarding the correlation of clinical disease symptoms and MRI findings, four of 31 patients (12.9%) who had the combination of intestinal dilatation, fold abnormalities, bowel wall thickening, and lymphadenopathy on MRI also demonstrated clinical signs and symptoms, including severe diarrhea and anemia. On the other hand, four of 31 patients (12.9%) who had an unremarkable MR examination, including no fold abnormalities, had only mild abdominal symptoms and mild anemia. The number of patients was too small to achieve statistical significance. MRI demonstrated a specificity and accuracy of 100%, and sensitivity of 79% and 75% for increased number of ileal folds and reversed fold pattern abnormality. Figure 3. Coronal T2-weighted HASTE image in a 35-year-old patient, with celiac disease showing a jejunoileal fold pattern reversal. Note the decreased number of jejunal folds (arrow) and the increased number of ileal folds (arrowheads) in the dilated ileum. Note also the dilatation of small bowel loops with no evidence of stenosis. Jejunal intussusception is apparent in two different locations, as shown by the target sign (open arrows). plane (which allows better distinction between the jejunum and ileum), higher intrinsic soft-tissue contrast resolution, and the lack of ionizing radiation in patients who may require serial follow-up studies. Prior studies in patients with abdominal malignancies screened for extrahepatic disease found that MRI was superior to CT DISCUSSION The current study shows that MRI is able to identify bowel and extraintestinal findings of celiac disease in adult patients. The ability to visualize a variety of smallbowel diseases with MRI was previously described (14,15,21 23). CT is a good modality for visualizing the small bowel. However, MRI has certain advantages over CT, including direct acquisition of images in the coronal Figure 4. Axial T2-weighted HASTE image in a 41-year-old patient with celiac disease, showing ileal jejunization with diffuse moderate bowel wall thickening (arrowheads).

4 628 Tomei et al. Figure 5. Axial T2-weighted HASTE image in a 36-year-old patient with endoscopically proven duodenal extrabulbar ulcers. Note the duodenal focal wall thickening (arrow) along with luminal narrowing and prestenotic dilatation. in detecting small-bowel lesions and lymphadenopathy (12,21 23). A comparison with the new multidetector CT would have to be performed to determine the relative merits of each modality from a diagnostic standpoint. Nonetheless, the lack of radiation will remain an important advantage of MRI. Barium studies are an important tool for diagnosing intestinal disease, and enteroclysis demonstrates excellent diagnostic accuracy; however, a complementary cross-sectional imaging method is also useful in most patients with celiac disease to identify the full extent of abnormalities (6). Extraintestinal abnormalities, such as lymphadenopathy, are commonly seen in the setting of celiac disease and are an important observation that may effect clinical management. Based on the results of this and previous work, at one of our institutions abdominal MRI is becoming the first-choice imaging method and is often the only modality performed in patients with suspected celiac disease. Bowel dilatation was a common finding in our study; however, it is a nonspecific feature and can be seen in many intestinal diseases, including self-limiting acute diseases. Whipple disease and Crohn disease may also show bowel dilatation, lymphadenopathy, and bowel wall thickening. Intestinal scleroderma may present with the malabsorption syndrome, and the small bowel is typically dilated and abnormally fluid-filled (1,5,24,25). Bowel dilatation may also simulate bowel obstruction, paralytic ileus, or uncommon entities such as intestinal scleroderma, and the patient s clinical history is crucial for determining the likely processes. Nonetheless, we believe that in the absence of imaging findings or a clinical history consistent with obstructive disease, drug-induced hypotonia, or acute bowel infection, celiac disease should be considered in patients with bowel dilatation and no underlying cause. The fold pattern abnormalities of celiac disease, including jejunoileal fold pattern reversal and jejunization of the ileum, have been observed on barium studies (6) and CT (9,13). In our series, jejunization of the ileum was observed more frequently than jejunoileal fold pattern reversal in celiac patients (48.4% (15/31) and 38.7% (12/31), respectively). A sizeable minority of patients (12.9% (4/31)) showed a normal fold pattern. Mural thickening in the setting of celiac disease may reflect submucosal edema and varying degrees of inflammation. At the present time, the precise cause of bowel wall thickening is complex and, in the majority of cases, incompletely understood (2,6,10,15). The presence of duodenum abnormalities, such as dilatation, strictures, or stenosis due to peptic ulcer disease, in patients with celiac disease was previously described (26). Our findings concur with prior reports that described adequate depiction of duodenal disease on MRI (27). Transient intussusception is a common finding in adult patients with celiac disease (10) and is well demonstrated by MR examination (19,20). Therefore, the presence of intussusception in an adult patient with no underlying cause should raise the suspicion of celiac disease. In our series, benign mesenteric lymphadenopathy was the most common extraintestinal finding (41.9%; 13/31). Because celiac disease is a risk factor for the development of intestinal malignancy (28), particularly lymphoma (29,30), patients with lymph node enlargement should be followed closely. The diagnosis of lymphoma in patients with unexplained or relapsing symp- Figure 6. Coronal True-FISP image in a 38-year-old patient with celiac disease that demonstrates an increased caliber of mesenteric vessels (vascular engorgement) and the presence of multiple bulky lymph nodes (arrow) along the mesenteric vessels.

5 MRI in Adult Celiac Disease 629 toms is difficult to establish, but must be excluded in these patients (31 33). MRI can adequately demonstrate nodal enlargement. Mesenteric vascular changes have been observed by ultrasound (34). At the present time, there are no standards to quantify mesenteric vascular abnormalities on MRI; therefore, this is currently a subjective evaluation. In our experience, vessels along the mid-length of the mesentery that are 3 mm in diameter are usually abnormal; however, we have not directly evaluated this in a study setting. All of the patients with small-volume ascites also demonstrated bowel wall thickening, vascular engorgement, and lymphadenopathy. The cause of ascites in the setting of celiac disease has not yet been elucidated. This study is limited by the small number of patients, which reflects the single-institution nature of this project. In addition, since the MR studies were obtained in symptomatic patients, it is possible that patients with latent and subclinical diseases were missed. The inclusion of only patients with symptoms may have led to a selection bias in our study. However, selection bias is invariably present in imaging studies, and we do not think that our final result would be affected by the introduction of this bias. Finally, of all of the MR features described above, only fold abnormalities should be considered a specific finding. We observed in this study that nonspecific findings on MRI may also reflect the presence of celiac disease. In conclusion, MRI is able to demonstrate bowel and extraluminal findings in adult patients with celiac disease. It is important to recognize the spectrum of MR findings described in this study in order to increase awareness of the possible diagnosis of celiac disease, which can then be confirmed by EMA and biopsy. REFERENCES 1. Catassi C, Rattsch IM, Fabiani E, et al. Celiac disease in the year 2000: exploring the iceberg. Lancet 1994;343: Farrell RJ, Kelly CP. Celiac sprue. N Engl J Med 2002;346: Bardella MT, Minoli G, Radaelli F, et al. Increased prevalence of celiac disease in patients with dyspepsia. Arch Int Med 2000;16: Dickey W, Hughes D. Prevalence of celiac disease and its endoscopic markers among patients having routine gastrointestinal endoscopy. Am J Gastroenterol 1999;94: Trier JS. Diagnosis of celiac sprue. Gastroenterology 1998;115: Lomoschitz F, Wschima E, Schober K, et al. Enteroclysis in adult celiac disease: diagnostic value of specific radiographic features. Eur Radiol 2003;13: Herlinger H, Maglinte DD. Jejunal fold separation in adult celiac disease: relevance of enteroclysis. Radiology 1986;158: Rettenbacher T, Hollerweger A, Macheiner P, et al. Adult celiac disease: US signs. Radiology 1999;211: Tomei E, Marini M, Messineo D, et al. Computed tomography of the small bowel in adult celiac disease: the jejunoileal fold pattern reversal. Eur Radiol 2000;10: Strobl PW, Warshauer DM. CT diagnosis of celiac disease. J Comput Assist Tomogr 1995;19: de Boer WA, Maas M, Tytgat GN. Disappearance of mesenteric lymphadenopathy with gluten free diet in celiac sprue. J Clin Gastroenterol 1993;16: Macari M, Hines J, Balthazar E, et al. Mesenteric adenitis: CT diagnosis of primary versus secondary causes, incidence and clinical significance in pediatric and adult patients. AJR Am J Roentgenol 2002;178: Tomei E, Diacinti D, Marini M, et al. Abdominal CT findings may suggest coeliac disease. Dig Liver Dis 2005;37: Laghi A, Paolantonio P, Catalano C, et al. MR imaging of the small bowel using polyethylene glycol as oral contrast agent in adults and children with celiac disease. AJR Am J Roentgenol 2003;180: Laghi A, Carbone I, Catalano C, et al. Polyethylene glycol solution as an oral contrast agent for MR imaging of the small bowel. AJR Am J Roentgenol 2001;177: Frager D, Medwid SW, Baer JW, et al. CT of small bowel obstruction: value in establishing the diagnosis and determining the degree and cause. AJR Am J Roentgenol 1994;162: Bova JG, Friedman AC, Weser E, et al. Adaptation of the ileum in nontropical sprue: reversal of jejunoileal fold pattern. AJR Am J Roentgenol 1985;144: Balthazar EJ. CT of the gastrointestinal tract: principles and interpretation. AJR Am J Roentgenol 1991;156: Warshauer DM, Lee JKT. Adult intussusception detected at CT or MR imaging: clinical-imaging correlation. Radiology 1999;212: Marcos HB, Semelka RC, Worawattanakul S. Adult intussusception: demonstration by current MR techniques. Magn Reson Imaging 1997;9: Lee JKT, Marcos HB, Semelka RC. MR imaging of the small bowel using the HASTE sequence. AJR Am J Roentgenol 1998;170: Semelka RC, Shoenut JP, Silverman R, et al. Bowel disease: prospective comparison of CT and 1.5-T pre- and postcontrast MR imaging with T1-weighted fat-suppressed and breath-hold FLASH sequences. J Magn Reson Imaging 1991;1: Laghi A, Borrelli O, Paolantonio P, et al. Contrast enhanced magnetic resonance imaging of the terminal ileum in children with Crohn s disease. Gut 2003;52: Hin H, Bird G, Fisher P, et al. Celiac disease in primary care: case finding study. BMJ 1999;318: Jennings JS, Howdle PD. Celiac disease. Curr Opin Gastroenterol 2001;17: Schweiger GD, Murray JA. Postbulbar duodenal ulceration and stenosis associated with celiac disease. Abdom Imaging 1998;23: Marcos HB, Semelka RC, Noone TC, et al. MRI of normal and abnormal duodenum using half-fourier single-shot RARE and gadolinium-enhanced spoiled gradient echo sequences. Magn Reson Imaging 1999;17: Low RN, Semelka RC, Worawattanakul S, et al. Extrahepatic abdominal imaging in patients with malignancy: comparison of MR imaging and helical CT, with subsequent surgical correlation. Radiology 1999;210: Al-Kawas FH, Murgo A, Foshag L, et al. Lymphadenopathy in celiac disease: not always a sign of lymphoma. Am J Gastroenterol 1988; 83: Green PH, Jabri B. Celiac disease and other precursors to smallbowel malignancy. Gastroenterol Clin North Am 2002;31: Perret RS, Borne J. Case 3: T-cell lymphoma of the small bowel developing in long-standing celiac disease. AJR Am J Roentgenol 1998;171: Egan LJ, Stevens FM, McCarthy CF. Celiac disease and T-cell lymphoma. New Engl J Med 1996;335: Meijer JW, Mulder CJ, Goerres MG, Boot H, Schweizer JJ. Coeliac disease and (extra) intestinal T-cell lymphomas: definition, diagnosis and treatment. Scand J Gastroenterol Suppl 2004;241: Giovagnorio F, Picarelli A, Di Giovambattista F, et al. Evaluation with Doppler sonography of mesenteric blood flow in celiac disease. AJR Am J Roentgenol 1998;171: